Date of Original Version
A new numerical procedure for the study of finite temperature quantumdynamics is developed. The method is based on the observation that the real and imaginary time dynamical data contain complementary types of information. Maximum entropy methods, based on a combination of real and imaginary time input data, are used to calculate the spectral densities associated with real time correlation functions. Model studies demonstrate that the inclusion of even modest amounts of short-time real time data significantly improves the quality of the resulting spectral densities over that achievable using either real time data or imaginary time data separately.
Kim, S., Doll, J. D., & Freeman, D. L. (1998). Dynamic Path Integral Methods: A Maximum Entropy Approach Based on the Combined Use of Real and Imaginary Time Quantum Monte Carlo Data. Journal of Chemical Physics, 108(10), 3871-3875. doi: 10.1063/1.475790
Available at: http://dx.doi.org/10.1063/1.475790